JP3820008B2 - Pressure control method for fluid catalytic cracker - Google Patents

Description

といったルールが保存してある。
また、この知識ルールベース部１０１は、各種の運転変数の制約条件の中で目標とする変数が最適になるように操作変数を変更していく。そして、操作変数の変化に応じて各運転変数の変化が算出され、これが制約条件内であれば、さらに最適化のため操作変数を変更する。
【００１６】
操作変数決定部１０２は、知識ルールベース部１０１における上記ルールを用い、現状の運転データおよび制約条件より目標が最適になるように操作変数を算出する。
この操作変数の値は、操作変数出力部１０３を通して制御部１１１に出力される。
【００１７】
なお、制約条件入力部１０４には、制約条件としての下流の装置制約，熱交換器能力，バルブポジション等が、また、目標条件としての熱交換器出口温度等が、条件判断計算によって入力される。
また、運転データ入力部１０５は、主分留塔３における各中間リフラックス系におけるリフラックス温度，下流装置熱交換温度等の各種運転データを入力する。
【００１８】
図４は、予測モデル設定部２００の詳細なブロック図を示したものである。
同図に示す予測モデル設定部２００は、運転データ集積部２０１と、予測モデルを作成する予測モデル作成部２０２と、操作変数決定部２０３と、操作変数出力部２０４と、制約条件（目標条件）入力部２０５と、目標値探索部２０６、及び運転データ入力部２０７とからなっている。
【００１９】
ここで、運転データ集積部２０１は、各リフラックス系におけるリフラックス温度，リフラックス流量等の過去の運転データを記憶集積している。また、予測モデル作成部２０２は、過去の集積してある運転データにもとづいて動的モデルを作成するもので、具体的には、第一及び第二の中間リフラックス系のリフラックス量を調整して主分留塔の塔頂リフラックス流量を最少に制御するための予測モデルを、次のように作成する。
【００２０】
【数１】

【００２１】
ここで、

である。
【００２２】
操作変数決定部２０３は、予測モデルにより現状運転データの操作変数を変更して、制約条件内で目標（条件）に最も適合した運転条件を決定する。
すなわち、操作変数の上下限値，制御変数の上下限値，操作変数の変更幅，目的関数（目標条件）等を条件に線形計画法を利用して最適解を求める。
このように、現状の運転状態に対し最適化するよう変更して算出した操作変数は、操作変数出力部２０４を通して制御部２１１に出力される。
なお、操作変数出力部２０４は操作変数の微少変化をプロセス側の制御部が受け付けないときは、積算プログラムにより複数回の微少出力を積算して制御部がが受け付ける出力とする。
【００２３】
制約条件入力部２０５には、制約条件及び目標条件として、たとえば、
▲１▼主分留塔留出物（燃料ガス，ガソリン，ＬＣＯ，ＨＣＯ，ＣＬＯ）の品質（純度）、
▲２▼一部留出物（ｅｘ．ガソリン）流量（最大となるようにする）、
▲３▼クーラー３１ａ，３１ｂの冷却側流体流量、
▲４▼燃料ガス圧力、
▲５▼主分留塔中の製品抜出段の液負荷，蒸気負荷、
▲６▼一部棚段間の温度差、
などが入力される。
【００２４】
目標値探索部２０６では、制約条件入力部２０５からの制約条件を監視し、制約にかかっていない場合には目標流量を下げながら最適目標流量を探索し、制約範囲内での最低流量を確保する。
【００２５】
また、運転データ入力部２０７は、主分留塔の塔頂リフラックス流量及び主分留塔の第一及び第二の中間リフラックス量とリフラックス温度等の各種運転データを入力する。
【００２６】
上記のように、知識ルールベース設定部１００及び制御部１１１からなる制御系では、主分留塔３のクーラー３１ａ，３１ｂ及び周囲の制約条件を監視して各中間リフラックス系のリフラックス温度を制御し、予測モデル設定部２００及び制御部２１１からなる制御系では、主分留塔の各中間リフラックス温度の変化を検知し、流量特性に与える影響を予測モデルで予測して各中間リフラックス量の制御を行なう。
これら各制御は、たとえば知識ルールベースによる制御を１時間の周期で、予測モデルによる制御を３分間の周期で行ない、かつこれら両制御を複合的に行なっている。
その結果、主分留塔３の低圧制御ひいては反応塔１の低圧制御の自動化が可能となり、流動接触分解装置全体の自動制御が可能となった。
なお、知識ルールベースによる制御の周期は、予測モデルによる制御の周期よりも長くすることが好ましく、たとえば１０倍〜２０倍とすると、相互操作による干渉を受けずに制御可能となり好適である。
【００２７】
【発明の効果】
以上のように、本発明によれば、流動接触分解装置の圧力制御を自動化することができ、定常運転時から原料切替えあるいは運転モード切替えなどの過渡状態のときまで、常に反応塔を低圧化することができる。
【図面の簡単な説明】
【図１】本発明の制御方法を実施するための概略装置構成図を示す。
【図２】本発明の制御方法を実施するための制御系ブロック図を示す。
【図３】図２における知識ルールベース設定部の詳細ブロック図を示す。
【図４】図における予測モデル設定部の詳細ブロック図を示す。
【符号の説明】
１ 反応塔
２ 再生塔
３ 主分留塔
３１ａ，３１ｂ クーラー
１００ 知識ルールベース設定部
１０１ 知識ルールベース部
１０２ 操作変数決定部
１０３ 操作変数出力部
１０４ 制約条件入力部
１０５ 運転データ入力部
２００ 予測モデル設定部
２０１ 運転データ集積部
２０２ 予測モデル作成部
２０３ 操作変数決定部
２０４ 操作変数出力部
２０５ 制約条件入力部
２０６ 目標値探索部
２０７ 運転データ入力部[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a pressure control method for a fluid catalytic cracking apparatus for producing gasoline or the like.
[Prior art]
[0002]
The fluid catalytic cracking unit is a catalytic cracker of a petroleum fraction using a granular catalyst to produce a high octane gasoline, a light oil fraction, an LPG, and the like. As shown in FIG. 1, a reaction tower 1, a regeneration tower 2 and a main fraction are obtained. It consists of tower 3.
[0003]
In this fluid catalytic cracking apparatus, the heavy oil as a raw material is contacted with the catalyst at a high temperature in the reaction tower 1 and decomposed, and the decomposed oil uses the difference in boiling point in the main fractionation tower 3 to produce gasoline, light It is fractionated into light oil, heavy light oil, LPG and residual oil.
On the other hand, the catalyst used in the reaction tower 1 is regenerated and activated in the regeneration tower 2 and sent to the reaction tower 1 again using the pressure difference between the regeneration tower 2 and the reaction tower 1. Therefore, the pressure difference between the regeneration tower 2 and the reaction tower 1 needs to be kept constant while the pressure in the regeneration tower 2 is increased. The pressure in the reaction tower 1 is secondarily determined by the pressure in the main fractionator 3.
[0004]
Here, if the pressure in the reaction tower 1 is high, the oil decomposition rate deteriorates and the amount of catalyst input increases. Since the catalyst is expensive, in order to reduce the input amount of the catalyst as much as possible, it is necessary to keep the pressure in the reaction tower 1 as low as possible. For this purpose, it is necessary to reduce the pressure in the main fractionation tower 3. . In addition, the pressure in the regeneration tower 2 is lowered by lowering the pressure in the reaction tower 1, thereby lowering the temperature of the regeneration tower and reducing the amount of coke produced, thereby preventing catalyst deterioration.
On the other hand, in order to lower the pressure in the main fractionation tower 3, it is necessary to keep the reflux flow rate at the top of the tower, which moves equivalent to the pressure, as small as possible within the constrained range. The amount of flux corresponds to the reflux temperature, and the reflux temperature of each fraction is adjusted by adjusting the outlet temperature of each cooler.
[0005]
[Problems to be solved by the invention]
However, in order to control the main fractionator to a low pressure state,
First, since there is a restriction on the cooling capacity of each reflux system (excluding the tower reflux, the same applies hereinafter), it is necessary to operate while monitoring many restrictions on the opening of the bypass valve.
Secondly, since each time delay from the operation of each reflux amount to the influence of the flow rate at the top of the tower to be controlled is different, it is necessary to adjust in consideration of the time delay, The dynamics of the tower top reflux flow rate is nonlinear and has a large time delay.
Thirdly, since the reflux flow rate and the reflux temperature, which are the operation amounts, interfere with each other, it is necessary to perform an operation in consideration of the influence degree and the delay time.
Fourthly, the amount of heavy heavy oil that is a raw material depends mainly on the amount of production from upstream equipment and is not constant. Therefore, the processing amount and raw material properties change every moment, and the reflux flow rate at the top of the tower changes. Therefore, in order to reduce the pressure while stabilizing the tower top reflux flow rate, it is necessary to operate each reflux amount in a well-balanced manner while constantly monitoring the overall heat balance.
In the actual site, it was difficult to fully automate the pressure control of the main fractionation tower because it was necessary to cope with the experience of an experienced operator and PID control.
[0006]
The present invention has been considered in view of the above circumstances, and an object of the present invention is to provide a pressure control method for a fluid catalytic cracking apparatus that enables complete automatic control of the fluid catalytic cracking apparatus, which has been impossible until now.
[0007]
[Means for Solving the Problems]
In order to achieve the above object, the present invention performs the PID control in the conventional part as it is, performs the PID control as it is, replaces the part that has conventionally relied on the experience of the skilled operator with a knowledge rule base system, and performs monitoring and monitoring. The control of the fluid catalytic cracking apparatus is automatically performed by using the predictive model control together with the control of the difficult part even for the skilled operator with many operation points and performing the composite control as a whole.
[0008]
Specifically, the invention of claim 1 is a fluid catalytic cracking apparatus having a reaction tower, a regeneration tower, and a main fractionation tower, wherein the reflux temperature of the intermediate reflux system in the main fractionation tower is based on a knowledge rule base. And reducing the pressure of the main fractionation column by controlling the reflux flow rate at the top of the main fractionation column at a cycle shorter than the control based on the knowledge rule base based on the prediction model, and This is a method for lowering the pressure in the reaction tower by lowering the pressure in the main fractionation tower.
In this way, the pressure in the reaction tower is lowered, the oil decomposition rate is improved, the regeneration tower pressure and temperature are decreased, the amount of coke produced is reduced, and catalyst deterioration can be prevented. Improvements can be made.
[0009]
In this case, the control based on the prediction model is performed in a shorter cycle than the control based on the knowledge rule base. In other words, the control based on the knowledge rule base has a longer cycle than the control based on the prediction model.
Control based on the knowledge rule base is suitable for a system that has a slow response time and cannot be modeled, and control based on a prediction model is suitable for a system that can be modeled with a fast response time.
As described above, when the control of the system having a slow response time is performed in a longer cycle than the control of the system having a fast response time, mutual interference can be reduced and stable composite control can be performed.
[0010]
In the invention described in claim 2, the knowledge rule base unit is a rule obtained by making a procedure including monitoring and operation performed by a skilled operator into a rule by an IF-THEN control rule.
If the IF-THEN control rules are used in this way, the same operations including monitoring and operation as those performed by skilled operators can be automatically and reliably performed.
[0011]
According to a third aspect of the present invention, the knowledge rule base is determined by searching for an optimum outlet temperature setting value of a cooler that controls the temperature of the intermediate reflux oil in the main fractionator.
[0012]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described.
First, an outline of the embodiment will be described.
The cooler 31a determines the temperature of the first intermediate reflux oil in the main fractionator 3 in the fluid catalytic cracking apparatus shown in FIG. 1, and determines the temperature of the second intermediate reflux oil. Is the cooler 31b, and the amount of heat exchange in these coolers 31a and 31b is determined by downstream apparatus restrictions and heat exchange capacity restrictions. Each knowledge rule base setting unit 100 searches and determines the optimum outlet temperature setting value of each cooler 31a, 31b so as to satisfy this restriction and recover the maximum amount of heat.
On the other hand, the flow rate of the first and second reflux oils at the optimum temperature is adjusted by the adjustment valves 32a and 32b based on the calculation result in the prediction model setting unit 200, so that the optimum tower is within the restricted range. The top reflux flow rate is searched, and the low pressure in the main fractionation tower 3 and the reaction tower 1 is realized.
[0013]
FIG. 2 shows a block diagram of the control system in FIG.
In the figure, the control unit 111 opens the control valves (operation units) 131a and 131b of the coolers 31a and 31b based on the difference between the output value from the knowledge rule base setting unit 100 and the fed-back intermediate reflux temperature. The degree is PID controlled. Further, the control unit 211 performs PID control on the opening degree of the adjustment valves (operation units) 32a and 32b based on the difference between the output value from the prediction model setting unit 200 and the fed back intermediate reflux amount.
[0014]
FIG. 3 is a detailed block diagram of the knowledge rule base setting unit 100.
The knowledge rule base setting unit 100 shown in the figure includes a knowledge rule base unit 101, an operation variable determination unit 102, an operation variable output unit 103, a constraint condition (target condition) input unit 104, and an operation data input unit 105. It is made up of.
[0015]
Here, the knowledge rule base unit 101 stores the monitoring and operation procedures performed by the skilled operator as rules according to IF-THEN control rules.
For example, “Monitor the constraining conditions around the cooler, and if the constraint is not met, search for the optimum outlet temperature while lowering the target exit temperature and recover the maximum heat. If the constraint is met, perform the return operation. In order to optimize within the constraints, such as "Avoid constraints."

Such a rule is saved.
In addition, the knowledge rule base unit 101 changes the operation variable so that the target variable is optimal among the constraint conditions of various operation variables. Then, the change of each operation variable is calculated according to the change of the operation variable, and if this is within the constraint condition, the operation variable is further changed for optimization.
[0016]
The operation variable determination unit 102 uses the above rules in the knowledge rule base unit 101 to calculate an operation variable so that the target is optimized based on current operation data and constraint conditions.
The value of the operation variable is output to the control unit 111 through the operation variable output unit 103.
[0017]
The constraint condition input unit 104 receives downstream apparatus constraints, heat exchanger capacity, valve position, etc. as constraint conditions, and heat exchanger outlet temperature, etc., as target conditions by condition judgment calculation. .
The operation data input unit 105 inputs various operation data such as a reflux temperature in each intermediate reflux system in the main fractionator 3 and a downstream device heat exchange temperature.
[0018]
FIG. 4 shows a detailed block diagram of the prediction model setting unit 200.
The prediction model setting unit 200 shown in the figure includes an operation data accumulation unit 201, a prediction model creation unit 202 that creates a prediction model, an operation variable determination unit 203, an operation variable output unit 204, and constraint conditions (target conditions). The input unit 205 includes a target value search unit 206 and an operation data input unit 207.
[0019]
Here, the operation data accumulation unit 201 stores and accumulates past operation data such as a reflux temperature and a reflux flow rate in each reflux system. The prediction model creation unit 202 creates a dynamic model based on past accumulated operation data. Specifically, the prediction model creation unit 202 adjusts the amount of reflux of the first and second intermediate reflux systems. Then, a prediction model for minimizing the top reflux flow rate of the main fractionator is created as follows.
[0020]
[Expression 1]

[0021]
here,

It is.
[0022]
The operation variable determination unit 203 changes the operation variable of the current operation data based on the prediction model, and determines the operation condition that best matches the target (condition) within the constraint conditions.
That is, an optimal solution is obtained using linear programming on the condition of the upper and lower limits of the manipulated variable, the upper and lower limits of the control variable, the change width of the manipulated variable, the objective function (target condition), and the like.
In this way, the manipulated variable calculated to be optimized for the current driving state is output to the control unit 211 through the manipulated variable output unit 204.
When the process-side control unit does not accept a slight change in the operation variable, the operation variable output unit 204 accumulates a plurality of micro-outputs by the accumulation program and outputs the output as accepted by the control unit.
[0023]
In the constraint condition input unit 205, as a constraint condition and a target condition, for example,
(1) Quality (purity) of the main fractionator distillate (fuel gas, gasoline, LCO, HCO, CLO),
(2) Partial distillate (ex. Gasoline) flow rate (maximum),
(3) Cooling side fluid flow rate of the coolers 31a and 31b,
(4) Fuel gas pressure,
(5) Liquid load, steam load at the product extraction stage in the main fractionation tower,
(6) Temperature difference between some shelves,
Etc. are entered.
[0024]
The target value search unit 206 monitors the constraint condition from the constraint condition input unit 205, and if the constraint is not satisfied, searches for the optimum target flow rate while lowering the target flow rate, and secures the minimum flow rate within the constraint range. .
[0025]
The operation data input unit 207 inputs various operation data such as the top reflux flow rate of the main fractionation column, the first and second intermediate reflux amounts and the reflux temperature of the main fractionation column.
[0026]
As described above, in the control system including the knowledge rule base setting unit 100 and the control unit 111, the coolers 31a and 31b of the main fractionator 3 and the surrounding constraints are monitored, and the reflux temperature of each intermediate reflux system is determined. In the control system comprising the prediction model setting unit 200 and the control unit 211, the change in each intermediate reflux temperature of the main fractionation column is detected, and the influence on the flow rate characteristic is predicted by the prediction model, and each intermediate reflux is detected. Control the amount.
In each of these controls, for example, control based on a knowledge rule is performed at a cycle of 1 hour, control by a prediction model is performed at a cycle of 3 minutes, and both these controls are performed in combination.
As a result, it is possible to automate the low pressure control of the main fractionation tower 3 and thus the low pressure control of the reaction tower 1, and the whole fluid catalytic cracking apparatus can be automatically controlled.
The period of control based on the knowledge rule base is preferably longer than the period of control based on the prediction model. For example, if it is 10 to 20 times, the control can be performed without receiving interference due to the mutual operation.
[0027]
【The invention's effect】
As described above, according to the present invention, the pressure control of the fluid catalytic cracking apparatus can be automated, and the pressure of the reaction tower is always reduced from the steady operation to the transient state such as the raw material switching or the operation mode switching. be able to.
[Brief description of the drawings]
FIG. 1 shows a schematic apparatus configuration diagram for carrying out a control method of the present invention.
FIG. 2 shows a block diagram of a control system for carrying out the control method of the present invention.
FIG. 3 is a detailed block diagram of a knowledge rule base setting unit in FIG. 2;
FIG. 4 shows a detailed block diagram of a prediction model setting unit in the figure.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 Reaction tower 2 Regeneration tower 3 Main fractionator 31a, 31b Cooler 100 Knowledge rule base setting part 101 Knowledge rule base part 102 Operation variable determination part 103 Operation variable output part 104 Restriction condition input part 105 Operation data input part 200 Prediction model setting Unit 201 operation data accumulation unit 202 prediction model creation unit 203 operation variable determination unit 204 operation variable output unit 205 constraint condition input unit 206 target value search unit 207 operation data input unit

Claims (3)

In a fluid catalytic cracking apparatus having a reaction tower, a regeneration tower and a main fractionation tower,
Control of the reflux temperature of the intermediate reflux system in the main fractionator based on a knowledge rule base, and control of the tower top reflux flow rate in the main fractionator based on the knowledge rule base based on a prediction model The pressure in the main fractionator is lowered by controlling with a shorter period ,
And the pressure control method in the fluid catalytic cracking apparatus characterized by lowering the pressure of the reaction tower by lowering the pressure of the main fractionating tower and increasing the reaction efficiency.   2. The pressure control method for a fluid catalytic cracking apparatus according to claim 1, wherein the knowledge rule base is a rule obtained by making a procedure including monitoring and operation performed by a skilled operator into a rule using an IF-THEN control rule. The flow according to claim 1 or 2, wherein the knowledge rule base searches and determines an optimum outlet temperature setting value of a cooler that controls the temperature of the intermediate reflux oil in the main fractionator. Pressure control method for catalytic cracking device.

Images